This paper demonstrates the complementarity between the systems approach and artificial intelligence (AI) in managing complex real-world situations, based on the authors’ thirty years of joint work. The focus is on practical experiences, primarily in the Peruvian context, where Soft Systems Methodology (SSM) was integrated with AI-based multi-criteria decision-modeling approaches. The methodology is based on applying SSM stages, depending on objectives of the specific use case, and integrating AI-based methods into stages that require culturally feasible and systemically desirable assessment of situations and decision alternatives. The approach is illustrated with three use cases. First, the intelligent performance evaluation of 144 non-financial Peruvian public enterprises, providing a systematic and data-driven basis for identifying improvement opportunities. Second, the design of an intelligent decision room to address strategic management challenges for a major Peruvian private enterprise group. Third, the intelligent assessment and management of risks in the Peruvian energy and mining sectors, enabling proactive and informed risk mitigation strategies. These cases illustrate how combining systems thinking with AI techniques can transform unstructured problems into actionable insights, bridging the gap between qualitative understanding and quantitative analysis. The approach provides a robust, effective and efficient framework for addressing complexity, supporting strategic decision-making in an uncertain world, and improving performance in diverse organizational and sectoral contexts.

Abstract Operational decision support systems often rely on ordered risk states rather than calibrated probabilities and therefore require transparency and consistency. This paper introduces DEX‐LL , a data‐inductive extension of the Decision EXpert (DEX) qualitative decision‐support framework that constructs hierarchical ordinal decision models with explicit, monotone decision tables. Intermediate decision concepts are learned by aggregating small subsets of categorical attributes, with monotonicity enforced as a semantic requirement via isotonic projection on discrete ordinal grids. To stabilize rule induction under rare events, conservative uncertainty‐adjusted scoring based on Wilson lower confidence bounds is employed. Large language models are used only to propose candidate attribute groupings for intermediate concepts, while all rule induction and validation remain fully data‐driven and deterministic. Evaluation on six seasons of ski resort operational data shows that DEX‐LL yields interpretable, auditable, and action‐consistent risk stratifications that remain competitive with standard predictive models when evaluated on an operationally aligned ordinal decision scale.

This research investigates the variability of decision-making preferences, represented in terms of decision rules and criteria weights, in the context of the qualitative multi-criteria method DEX (Decision EXpert). We study the differences between decision rules acquired from different subjects (inter-personal differences) and from the same subjects at different times (intra-personal differences). We also assess the consistency of so-acquired rules and the ability of subjects to estimate the importance (weights) of criteria. The methodological approach consisted of two surveys among students, carried out about one and a half month apart. Four thematic areas were addressed in the questionnaires: selection of study programs, student success, car purchase decisions, and choices regarding everyday shopping venues. In both survey periods, participants were required to assess the importance of these criteria and to define decision rules according to the DEX method. The findings provide insights into the stability of decision-making processes among participants and in time. The results indicate a high variability of decision rules, both inter- and intra-personal. Intra-personal drift is lower than inter-personal differences, but not by much (three-quarters of the latter). The consistency of rules varied between small decision tables with clearly ordered criteria, where it was almost perfect, and large decision tables with less apparent preferential relations. Defining fully consistent decision tables turned out to be hard, indicating the need for automated consistencychecking tools. Criteria weights also drifted in time at the rate about 9% (user-provided weights) and 10-27% (weights assessed algorithmically from decision rules). The main contributions of this study are identified and quantified magnitudes of decision rules variability and consistency.

The development of decision support systems (DSS) for agriculture increasingly relies on complex decision models, yet transforming such models into operational, user-friendly software remains challenging. DEXiWare is a software framework designed to support the development of web-based, cooperative DSS based on decision models built with the DEX (Decision EXpert) method. The framework provides a standardized workflow for operationalizing decision models, including automated model import, data handling, assessment, and scenario analysis, within a reusable backend–frontend architecture. DEXiWare integrates backend services, a web-based user interface, and a decision engine supporting top-down (goal-seeking) and bottom-up (what-if) scenario exploration. The framework is evaluated through its application in multiple agricultural DSS and through usability testing with stakeholders, demonstrating its applicability for translating qualitative decision models into operational decision support tools for sustainability assessment in agricultural production systems.

The concept of sustainable mobility is aimed at minimising environmental impacts of transportation systems while meeting the needs of individuals and communities. This includes encouraging citizens to choose sustainable modes of transportation: walking, cycling, public transport, carpooling, and telecommuting. We present an approach at rewarding organisations that actively support the sustainable mobility of their employees, and propose a framework for awarding a sustainable mobility certificate to organisations that fulfil sustainable mobility goals and objectives. The assessment is carried out using a qualitative rule-based multi-criteria model, which considers 50 indicators. Other elements of the certification process include methods for assessing the mobility structure of employees in the organisation and its potential for improvement. In this paper, we present the main components of the proposed certification framework and illustrate its application for assessing the status of sustainable mobility of employees at a Slovenian research institute.

We experimentally assessed the capabilities of two mainstream artificial intelligence chatbots, ChatGPT and DeepSeek, to support the multi-criteria decision-making process. Specifically, we focused on using the method DEX (Decision EXpert) and investigated their performance in all stages of DEX model development and utilization. The results indicate that these tools may substantially contribute in the difficult stages of collecting and structuring decision criteria, and collecting data about decision alternatives. However, at the current stage of development, the support for the whole multi-criteria decision-making process is still lacking, mainly due to occasionally inconsistent and erroneous execution of methodological steps. To leverage the strengths of both approaches, we also propose a hybrid workflow for DEX model development that begins in the LLM and continues in the specialized DEXiWin software.

The potential of large language models (LLMs) was examined in the context of decision tables as used in the qualitative multi-criteria decision modelling method DEX (Decision EXpert). Interactive dialogues were conducted with two open-source LLM chatbots, DeepSeek and Llama, with their outputs evaluated from the perspective of an expert decision analyst. The interaction focused on the construction and modification of a single decision table, as well as its interpretation in terms of table properties, decision rules, attribute weights, and visualizations. Findings suggest that LLMs offer a convenient means for learning and executing simple tasks. However, at their current stage of development, they remain inconsistent and prone to errors, rendering them unsuitable for serious applications.

DEX (Decision EXpert) is a qualitative multi-criteria decision modelling method, and DEXi is software that supports the development of DEX models and their use for the evaluation and analysis of decision alternatives. We present DEXi Suite, a new generation of DEXi software, aimed at replacing the reliable and trusted, but dated, DEXi Classic software. DEXi Suite has been designed to employ a more modern and flexible software architecture and support extended functionality, while still remaining user-friendly and free to use. Currently, DEXi Suite consists of a DEXi modelling class library (DEXiLibrary), interactive desktop model editing and decision analysis software (DEXiWin) and a command-line evaluator (DEXiEval).

The focus of this study is to integrate the DEX (Decision EXpert) decision-modeling method in architectural and urban design (A & UD) competitions. This study aims to assess the effectiveness of integrating the DEX (Decision EXpert) decision-modeling method into the evaluation process of A & UD competitions to enhance decision-making transparency, objectivity, and efficiency. By using symbolic values in decision models, the approach offers a more user-friendly alternative to the conventional jury decision-making process. The practical application of the DEX method is demonstrated in the Rhinoceros 3D environment to show its effectiveness in evaluating A & UD competition project solutions related to the development of the smart city. The results indicate that the DEX method, with its hierarchical and symbolic values, significantly improves the simplicity of the evaluation process in A & UD competitions, aligning it with the objectives of the smart cities. This method provides an efficient, accessible, and viable alternative to other multi-criteria decision-making approaches. This study importantly contributes to the field of architectural decision making by merging qualitative multi-criteria decision models into the CAD environment, thus supporting more informed, objective, and transparent decision-making processes in the planning and development of smart cities.

As urban populations rise globally, cities face increasing challenges in managing urban mobility. This paper addresses the question of identifying which modifications to introduce regarding city mobility by evaluating potential solutions using city-specific, subjective multi-objective criteria. The innovative AI-based recommendation engine assists city planners and policymakers in prioritizing key urban mobility aspects for effective policy proposals. By leveraging multi-criteria decision analysis (MCDA) and ±1/2 analysis, this engine provides a structured approach to systematically and simultaneously navigate the complexities of urban mobility planning. The proposed approach aims to provide an open-source interoperable prototype for all smart cities to utilize such recommendation systems routinely, fostering efficient, sustainable, and forward-thinking urban mobility strategies. Case studies from four European cities—Helsinki (tunnel traffic), Amsterdam (bicycle traffic for a new city quarter), Messina (adding another bus line), and Bilbao (optimal timing for closing the city center)—highlight the engine’s transformative potential in shaping urban mobility policies. Ultimately, this contributes to more livable and resilient urban environments, based on advanced urban mobility management.

In this paper, we present an approach for the assessment of sustainable mobility of employees that is designed to be used in a unified way in different contexts. The approach is intended for assessment and self-assessment of organisations, for supporting decisions regarding sustainable mobility activities and eventually as the basis for awarding organisations with a sustainable mobility certificate. Its main contributions are tailored criteria and parameters for exposing sustainable mobility characteristics of an organisation in terms of current situation and future potential. The paper provides a detailed description and explanation of the methodology and an example of an application in practice. The results indicate that the approach is viable and operational and that the assessment results well represent the situation and provide clear indications of the challenges and the paths to improvement.

DEX (Decision EXpert) is a qualitative multi-criteria decision-modelling method in which decision alternatives are evaluated according to decision rules, elicited from individual decision makers. In this preliminary study, we assessed the differences between decision rules acquired from different subjects (inter-personal differences) and from the same subjects at different times (intra-personal differences). We also assessed the consistency of so-acquired rules and the ability of subjects to estimate the importance (weights) of criteria. The results indicate a high variability of decision rules, both inter-and intra-personal. Intra-personal drift is lower than inter-personal differences, but not substantially. The consistency of rules varied between a small decision table with clearly ordered criteria, where it was almost perfect, and a large decision table with less apparent preferential relations, where it was rather poor at the average level of 0.77. Criteria weighs also drifted at the rate 9-19% per month.

Creating decision models for risk assessment of ski injuries is a challenging task. Ski injuries are rare events, but they carry a high cost, that is, can cause working or movement disabilities. Usually, ski risk assessment is performed on small-scale, case-controlled studies where the effect of a single factor is evaluated. Recently, data mining and machine learning algorithms are being employed for ski risk assessment and injury prediction. However, these models do not generally satisfy the need for interpretation of the decision model, do not provide explanations for the predictions, and in general do not ensure the completeness and consistency of decision rules. To make data mining and machine learning models useful, one needs to implement the aforementioned properties. Decision support systems are expected to have these properties; however, the process of building such decision support systems is still tedious: it has to consider human biases, assumptions, and subjective values, as well as focus on the decision problem being solved. We propose a method for extraction of decision models from data at hand. Our method DIDEX, Data Induced DEcision eXpert, builds models that have desirable properties for inclusion in decision support systems. The proposed method is used to build a decision model for ski injury prediction based on data from Mt. Kopaonik ski resort, Serbia. The results show that DIDEX generates up to a five times simpler model compared to the existing domain expert DEX models while having a 6% better predictive accuracy. Additionally, its predictive accuracy is comparable to similar machine learning algorithms, such as decision tree classifiers, random forest, and logistic regression.

In the project NARSIS – New Approach to Reactor Safety ImprovementS – possible advances in safety assessment of nuclear power plants (NPPs) were considered, which also included possible improvements in the field of management of low probability accident scenarios. As a part of it, a supporting software tool for making decisions under severe accident management was developed. The mentioned tool, named Severa, is a prototype demonstration-level decision supporting system, aimed for the use by the technical support center (TSC) while managing a severe accident, or for the training purposes. Severa interprets, stores and monitors key physical measurements during accident sequence progression. It assesses the current state of physical barriers: core, reactor coolant system, reactor pressure vessel and containment. The tool gives predictions regarding accident progression in the case that no action is taken by the TSC. It provides a list of possible recovery strategies and courses of action. The applicability and feasibility of possible action courses in the given situation are addressed. For each action course, Severa assesses consequences in terms of probability of the containment failure and estimated time window for failure. At the end, Severa evaluates and ranks the feasible actions, providing recommendations for the TSC. The verification and validation of Severa has been performed in the project and is also described in this paper. Although largely simplified in its current state, Severa successfully demonstrated its potential for supporting accident management and pointed toward the next steps needed with regard to further advancements in this field.

In a context of recall of glyphosate and questioning about the intensive use of ploughing, tools are needed to offer an alternative for perennial weed control. Perennial weeds are a major problem in production fields, organic or conventional farming. To control them, a system approach is required. This is the aim of the work package Modelling of the European project AC/DC-weeds. Multi-attribute qualitative modelling, performed thanks to IPSIM (Injury Profile SIMulator), enables the evaluation of weed infestation for three perennial weeds included one Cirsium arvense. The models consider the effect of weather, soil and cropping practices, and their interaction on weed infestation. The model outputs were confronted to independent field observations collected across 6 fields, over a 16- year period in 3 sites. IPSIM-Cirsium showed a satisfactory predictive quality (accuracy of 78.2%) IPSIM-Cirsium can be used as a tool for crop advisors and researchers to assist the design of systems less reliant on herbicides, for farmers and advisers to assess ex-ante prototypes of cropping systems, and for teachers as an educational tool to share agroecological weed management knowledge.

The international community has come a long way in developing a consensus that the remediation and management of naturally occurring radioactive materials and nuclear legacy sites will benefit from the use of the framework for risk-informed decision-making. Such a framework should ideally integrate risk assessment and decision-making. The framework presented in this paper specifically addresses the needs and expectations in the wider socio-economic and environmental context, as well as a narrower human health context. The framework was demonstrated as part of the International Atomic Energy Agency's second Modelling and Data for Radiological Impact Assessments Programme. Three case studies, which have used or could use this integrative approach, are used for illustration. The first concerns remediation from uranium mining activities at Beaverlodge Lake in northern Saskatchewan, Canada, engaging stakeholders (also called 'interested parties') in the decision-making process on further options. The second case study suggests how decision analysis could support the selection of the best option for waste disposal for uranium ore processing at Žirovski vrh, Slovenia, taking into account a potential landslide and migration of waste throughout the adjacent valley in the event of flooding. The third case study presents the process and results of radiological safety assessment of the Kepkensberg sludge basin in Tessenderlo area, Belgium both before and after the disposal of material from remediation of the nearby Winterbeek River. It illustrates how such assessments could interface with decision analysis for the purpose of supporting the regulatory decisions related to future approval of a waste disposal option. Results show that formal stakeholder engagement in decision analysis provides a strong contribution to objective, robust, and transparent decision-making not only for radiation protection area but also in others where health and environmental impacts are of concern. A number of recommendations for future work have also been made.

Multi-attribute decision analysis is an approach to decision support in which decision alternatives are evaluated by multi-criteria models. An advanced feature of decision support models is the possibility to search for new alternatives that satisfy certain conditions. This task is important for practical decision support; however, the related work on generating alternatives for qualitative multi-attribute decision models is quite scarce. In this paper, we introduce Bayesian Alternative Generator for Decision Support Models (BAG-DSM), a method to address the problem of generating alternatives. More specifically, given a multi-attribute hierarchical model and an alternative representing the initial state, the goal is to generate alternatives that demand the least change in the provided alternative to obtain a desirable outcome. The brute force approach has exponential time complexity and has prohibitively long execution times, even for moderately sized models. BAG-DSM avoids these problems by using a Bayesian optimization approach adapted to qualitative DEX models. BAG-DSM was extensively evaluated and compared to a baseline method on 43 different DEX decision models with varying complexity, e.g., different depth and attribute importance. The comparison was performed with respect to: the time to obtain the first appropriate alternative, the number of generated alternatives, and the number of attribute changes required to reach the generated alternatives. BAG-DSM outperforms the baseline in all of the experiments by a large margin. Additionally, the evaluation confirms BAG-DSM's suitability for the task, i.e., on average, it generates at least one appropriate alternative within two seconds. The relation between the depth of the multi-attribute hierarchical models-a parameter that increases the search space exponentially-and the time to obtaining the first appropriate alternative was linear and not exponential, by which BAG-DSM's scalability is empirically confirmed.

DEX (Decision EXpert) is a hierarchical, qualitative, rule-based, multi-criteria decision modeling method. It combines multi criteria decision analysis with artificial intelligence and is particularly suited for sorting/classification decision problems. DEX puts special attention on the transparency, comprehensibility, consistency, and completeness of decision models, as well as on methods for the analysis, justification, and explanation of decisions. The approach relies on using software tools that actively support the decision maker in both the creation and utilization stages of the process. Since its inception in the 1980s, DEX has been successfully applied in hundreds of real-world decision projects in various areas, including economy, ecology, agronomy, medicine, and health care. In the last decade, there is an increasing trend of including DEX models in decision support systems. In this chapter, DEX is described from the theoretical and practical viewpoint and further explained in terms of motivation, history, software, applications, and method extensions. The presentation is supported by three examples: a didactic example of employee selection and two real-world industrial applications of choosing a raw-material location and assessing electric energy production technologies, respectively.

Background Congestive heart failure (CHF) is a disease that requires complex management involving multiple medications, exercise, and lifestyle changes. It mainly affects older patients with depression and anxiety, who commonly find management difficult. Existing mobile apps supporting the self-management of CHF have limited features and are inadequately validated. Objective The HeartMan project aims to develop a personal health system that would comprehensively address CHF self-management by using sensing devices and artificial intelligence methods. This paper presents the design of the system and reports on the accuracy of its patient-monitoring methods, overall effectiveness, and patient perceptions. Methods A mobile app was developed as the core of the HeartMan system, and the app was connected to a custom wristband and cloud services. The system features machine learning methods for patient monitoring: continuous blood pressure (BP) estimation, physical activity monitoring, and psychological profile recognition. These methods feed a decision support system that provides recommendations on physical health and psychological support. The system was designed using a human-centered methodology involving the patients throughout development. It was evaluated in a proof-of-concept trial with 56 patients. Results Fairly high accuracy of the patient-monitoring methods was observed. The mean absolute error of BP estimation was 9.0 mm Hg for systolic BP and 7.0 mm Hg for diastolic BP. The accuracy of psychological profile detection was 88.6%. The F-measure for physical activity recognition was 71%. The proof-of-concept clinical trial in 56 patients showed that the HeartMan system significantly improved self-care behavior (P=.02), whereas depression and anxiety rates were significantly reduced (P<.001), as were perceived sexual problems (P=.01). According to the Unified Theory of Acceptance and Use of Technology questionnaire, a positive attitude toward HeartMan was seen among end users, resulting in increased awareness, self-monitoring, and empowerment. Conclusions The HeartMan project combined a range of advanced technologies with human-centered design to develop a complex system that was shown to help patients with CHF. More psychological than physical benefits were observed. Trial Registration ClinicalTrials.gov NCT03497871; https://clinicaltrials.gov/ct2/history/NCT03497871. International Registered Report Identifier (IRRID) RR2-10.1186/s12872-018-0921-2

This study tested the effectiveness of HeartMan—a mobile personal health system offering decisional support for management of congestive heart failure (CHF)—on health-related quality of life (HRQoL), self-management, exercise capacity, illness perception, mental and sexual health. A randomized controlled proof-of-concept trial (1:2 ratio of control:intervention) was set up with ambulatory CHF patients in stable condition in Belgium and Italy. Data were collected by means of a 6-min walking test and a number of standardized questionnaire instruments. A total of 56 (34 intervention and 22 control group) participants completed the study (77% male; mean age 63 years, sd 10.5). All depression and anxiety dimensions decreased in the intervention group (p < 0.001), while the need for sexual counselling decreased in the control group (p < 0.05). Although the group differences were not significant, self-care increased (p < 0.05), and sexual problems decreased (p < 0.05) in the intervention group only. No significant intervention effects were observed for HRQoL, self-care confidence, illness perception and exercise capacity. Overall, results of this proof-of-concept trial suggest that the HeartMan personal health system significantly improved mental and sexual health and self-care behaviour in CHF patients. These observations were in contrast to the lack of intervention effects on HRQoL, illness perception and exercise capacity.